The Carbon Cycle: Budgets, Trends, and Lessons from Southern - - PowerPoint PPT Presentation

The Carbon Cycle: Budgets, Trends, and Lessons from Southern Hemisphere Measurements

• A. Modelling and Interpretation
• B. Baring Head surface CO2 data
• C. Lauder ground based remote sensing CO2 measurements
• D. Lauder surface CO2 data
• E. Rainbow Mountain surface CO2 data

Sara Mikaloff Fletcher1A, Vanessa Sherlock1ACD, Britt Stephens4ABDE, Gordon Brailsford1ABDE, Brian Connor3AC, John Robinson1C, Dan Smale1D, Peter Franz1E, Antony Gomez1BDE, Mike Kotkamp1D, Rowena Moss1BD, Katja Riedel1B, Hisako Shiona1C, Australian Collaborators: David Griffith2ACD, Nicholas Deutscher2*ACD, Ronald Macatangay2C, Martin Riggenbach2D, Clare Murphy2C, Nicholas Jones2C, Graham Kettlewell2C … and TCCON collaborators at CalTech, JPL, and KIT

1. NIWA, NZ; 2. University of Wollongong; 3. BC Consulting, NZ; 4. National Center for Atmospheric Research (NCAR), USA * Now at University of Bremen, Inst. Of Environmental Physics.

Overview

• Regional and global CO2 fluxes to the

atmosphere: overview and key science questions

• Using atmospheric trace gas measurements to

infer regional carbon fluxes

– Surface measurements – New remote sensing data

• Case study: Southern Hemisphere surface and

remote sensing data as a new window onto biomass burning emissions

The Early Keeling Curve

Atmospheric CO2 at Mauna Loa, Hawaii

310 320 330 340 350 360 370 380 1958 1963 1968 1973 1978 1983 1988 1993 1998 2003 2008 CO2 (ppm)

Atmospheric CO2 at Mauna Loa, Hawaii

Global Carbon Project 2010; Updated from Le Quéré et al. 2009, Nature Geoscience; Canadell et al. 2007, PNAS

Human Perturbation of the Global Carbon Budget

Sink Source

Time (y)

5 10 10 5 1850 1900 1950 2000

1.1±0.7

deforestation

CO2 flux (PgC y-1)

2000-2009

(PgC) 1 Petagram of carbon (PgC) = 1 billion metric tons of carbon

Human Perturbation of the Global Carbon Budget Time (y)

Global Carbon Project 2010; Updated from Le Quéré et al. 2009, Nature Geoscience; Canadell et al. 2007, PNAS

5 10 10 5 1850 1900 1950 2000 deforestation fossil fuel emissions

Sink Source

CO2 flux (PgC y-1) 7.7±0.5 1.1±0.7

2000-2009

(PgC) 1 Petagram of carbon (PgC) = 1 billion metric tons of carbon

Human Perturbation of the Global Carbon Budget

Global Carbon Project 2010; Updated from Le Quéré et al. 2009, Nature Geoscience; Canadell et al. 2007, PNAS

5 10 10 5 1850 1900 1950 2000

4.1±0.1

fossil fuel emissions deforestation atmospheric CO2

Sink Source

Time (y)

CO2 flux (PgC y-1) 7.7±0.5 1.1±0.7

2000-2009

(PgC) 1 Petagram of carbon (PgC) = 1 billion metric tons of carbon

Human Perturbation of the Global Carbon Budget

Global Carbon Project 2010; Updated from Le Quéré et al. 2009, Nature Geoscience; Canadell et al. 2007, PNAS

5 10 10 5 1850 1900 1950 2000 atmospheric CO2 fossil fuel emissions deforestation

• cean

2.3±0.4

• cean

Sink Source

Time (y)

CO2 flux (PgC y-1)

(5 models)

4.1±0.1 7.7±0.5 1.1±0.7

2000-2009

(PgC) 1 Petagram of carbon (PgC) = 1 billion metric tons of carbon

Human Perturbation of the Global Carbon Budget

Global Carbon Project 2010; Updated from Le Quéré et al. 2009, Nature Geoscience; Canadell et al. 2007, PNAS

5 10 10 5 1850 1900 1950 2000

2000-2009

(PgC) atmospheric CO2

• cean

land fossil fuel emissions deforestation

(Residual)

Sink Source

Time (y)

CO2 flux (PgC y-1) 2.3±0.4

(5 models)

4.1±0.1 7.7±0.5 1.1±0.7 2.4

1 Petagram of carbon (PgC) = 1 billion metric tons of carbon

Inferring Fluxes from Observations

• Typically done using a network of ~100 surface sites
• Strongly limited by the observing network, especially in the

tropics and Southern Hemisphere

Figure Courtesy of WMO

Atmospheric Inversions

Key Questions

• What are the natural sources and sinks of CO2 to

the atmosphere? – Tropical and southern hemisphere regions particularly uncertain

• Can atmospheric measurements be used to

validate anthropogenic emissions reductions?

• What processes control variability and trends in

the natural fluxes?

• What does this imply for feedbacks between

climate change and the global carbon cycle?

New Zealand’s Greenhouse Gas Budget

• “In 2008, New Zealand’s total greenhouse gas emissions

were 74.7 million tonnes of carbon dioxide equivalent (Mt CO2-e), which means total emissions are now 13.9 Mt CO2-e (22.8%) higher than the 1990 level of 60.8 Mt CO2-e.”*

• “In 2008, net removals from afforestation, reforestation

and deforestation under the Kyoto Protocol were -14.4 Mt CO2-e.”*

• Atmospheric measurements and modeling provide an
• pportunity for independent, top-down verification of

carbon sequestration in forests

* New Zealand’s Greenhouse Gas Inventory 1990-2008 Ministry for the Environment, April 2010

Key Questions

• What are the natural sources and sinks of CO2 to

the atmosphere? – Tropical and southern hemisphere regions particularly uncertain

• Can atmospheric measurements be used to

validate anthropogenic emissions reductions?

• What processes control variability and trends in

the natural fluxes?

• What does this imply for feedbacks between

climate change and the global carbon cycle?

Key Questions

• What are the natural sources and sinks of CO2 to

the atmosphere? – Tropical and southern hemisphere regions particularly uncertain

• Can atmospheric measurements be used to

validate anthropogenic emissions reductions?

• What processes control variability and trends in

the natural fluxes?

• What does this imply for feedbacks between

climate change and the global carbon cycle?

Key Questions

• What are the natural sources and sinks of CO2 to

the atmosphere? – Tropical and southern hemisphere regions particularly uncertain

• Can atmospheric measurements be used to

validate anthropogenic emissions reductions?

• What processes control variability and trends in

the natural fluxes?

• What does this imply for feedbacks between

climate change and the global carbon cycle?

Contrasting Column and Surface Measurements

Southern Hemisphere TCCON stations

A Puzzle in the Southern Hemisphere Data

Detrended X CO2 (ppm)

 Column FTS

• Model

Surface In Situ CO2 at Darwin, AU

The Seasonal Cycle in the Column Observations

X CO2 (ppm) X CO2 (ppm)

CarbonTracker Tagged Tracer Simulations

• CarbonTracker fluxes optimized against the surface network

– Tagged forward simulations with optimized 2009 CT fluxes

• Separate tracer tags for:

– each of the 22 Transcom regions + AU/NZ split – fossil fuel, biomass burning, terrestrial biosphere, ocean flux

Contribution of Source Processes to the Column Seasonal Cycle

Model Observations Total Biosphere Burning Oceans Fossil Fuels

X CO2 (ppm) X CO2 (ppm)

Where Does the Biomass Burning Signal in the Column Data Come From?

How does this fit with what is known about biomass burning emissions?

• If the model-data mis-match is due to biomass burning

alone, it would imply that the South American+African biomass burning emissions are under-estimated by nearly a factor of two

• The GFED emissions are based on satellite data, and

generally considered to under-estimate biomass burning emissions

• CarbonTracker doesn’t optimise biomass burning

emissions

• However, other tracers associated with biomass burning

peak at Darwin and Lauder ~2 months earlier than the seasonal model-data mismatch in CO2…

Zonal Mean Biomass Burning Emissions From South America and Africa

Vertical Profile of South East Asian Biomass Burning Footprint at Darwin (ppm)

Could this be due to a bias in the model transport?

• Houweling et al. [2010] compared TCCON data to

four atmospheric models using the CT fluxes as boundary conditions

• They found similar seasonal biases at Darwin for

all of the models

• However, there could be biases common to all the

models or the the reanalysis fields forcing them

• Comparisons with aircraft data may provide a

degree of independent validation

Conclusions and Outlook

• Between 2000-2009, human beings emitted 7.7±0.5 PgC/yr to

the atmosphere from fossil fuel burning and cement production and another 1.1±0.7 PgC/yr from land use change

• The natural sinks took up over half of these emissions, with the
• cean absorbing 2.3±0.4 PgC/yr and the terrestrial biosphere

taking up 2.4 PgC/yr

• Model simulations suggest that the combination of surface and

column data in the Southern Hemisphere may provide a new window onto terrestrial fluxes from South America and Africa

• Future work will focus on

– Analysis of atmospheric CO2 simulations to understand observations – Assimilating new column and surface data using CarbonTracker

Thanks to:

• Andy Jacobson and the CarbonTracker-

North America Team

• For funding:

– NIWA: FRST, ISAT – UoW: ARC – NIES – NASA, CalTech